Pediat. Res. 10: 642-646 (1976) Adenosine deaminase (ADA) guanine triphosphate adenosine triphosphate severe combined immunodeficiency syndrome (SCID) cultured fibroblasts uridine Purine Dysfunction in Cells from Patients with Adenosine Deaminase Deficiency PAUL J. BENKE'37'AND DAVID DITTMAR Mailman Center and the Department of Pediarrics, University ofMiamiSchoo1 of Medicine, Miami, Florida, USA Extract METHODS Conversion of adenosine to inosine is decreased in adenosine Human fibroblast cell cultures were established from punch skin deaminase (ADA)-deficient fibroblasts at all concentrations of biopsies obtained from the forearm and grown in Gibco F12 adenosine tested. Adenosine is not differentially toxic to ADA-defi- medium (28) with 15% fetal calf serum and 1% penicillin-strep- cient fibroblasts except at very high (5 x 10-'-1 x 10-3M)adeno- tomycin solution. Three adenosine deaminase-deficient cell strains sine levels. Conversion of [14C]adenosineto GTP is not decreased in were obtained from other investigators (29). Cell cultures were ADA-deficient cells compared with control cell strains. Adenosine monitored for mycoplasma by cultivation in BBL Mvcop(asma conversion to ATP is the same as that in mutant cells except at broth base with enrichment (30) and none was found during the high nonphysiologic concentrations, at which it is slightly decreased course of the investigation. in ADA-deficient fibroblasts. This effect is probably not related to Fibroblasts were trypsinized, counted on a Coulter electronic the biochemical pathology of ADA-deficient lymphocytes in vivo. particle counter (31), and then plated at a density of 10,000 cells/cc Uridine, a pyrimidine compound, "rescues" control cells from the in wells in a tissue culture plate (32). Test medium was added after effects of adenosine toxicity, as previously reported, but it has no 20-24 hr. Fifteen percent horse serum (28) was used for toxicity protective effect on ADA-deficient fibroblasts. This suggests that studies because it does not contain endogenous ADA activity (22). uridine will have no therapeutic role in the treatment of the ADA- Fibroblasts were harvested by trypsinization just after reaching deficient form of severe combined immunodeficiency (SCID) confluency. They were washed three times in phosphate-buffered disease. saline before incubation with ['"]adenosine (33). General reaction conditions and procedures were similar to those previously de- Speculation scribed (3). After the incubation of cells with radioactive adeno- sine, the reaction was stopped by the addition of 25 4.2 M The purine base, hypoxanthine, may be necessary for normal HCI and the mixture was neutralized with 25 4.42 M KOH immunologic responses. and centrifuged. Twenty microliters of the supernatant were spotted on Whatman no. 3MM paper previously soaked in 0.05 M sodium borate, pH 9.0. Twenty nanomoles of carrier inosine, Severe combined immunodeficiency is an uncommon but deva- xanthine, guanosine, hypoxanthine, guanine, adenine, and adeno- stating disorder characterized by decreased ability to fight infec- sine were applied to the same spot and high voltage electrophore- tion and low or absent circulatory T and B lymphocytes.. Patients sis was carried out at 2,250 V for 2 hr. Spots were located under frequently die unless competent immunologic cells are successfully ultraviolet light, cut out, and counted in a Packard liquid transplanted. It has been known since 1972 that some of these scintillation counter with toluene-diphenyloxazole (PP0)-p-his- patients have almost complete absence of ADA activity (5, 13). In [2-(5-phenyloxazalyl)]-benzene(POPOP). a recent summary of the experience in the United States, Nucleotides were chromatographed by anion exchange on a Meuwissen er al. (16) found that 13 of 22 patients with SCID had thinlayer; PEI-cellulose plate (33) as previously described (17) ADA deficiency. ADA-deficient patients may be clinically and after a preparatory rinse in 200 ml MeOH-H,O (1: 1) for 20 min to pathologically distinguished from others with SCID because they remove purine bases, nucleosides, and salts (20). Nucleotide spots have characteristic radiologic abnormalities of the skelton and the were located, cut out, and counted as described above. presence of Hassall's corpuscles in the thymus (16. 26). Normal immunocompetent ce!!s are produced in ADA-deficient patients RESULTS for a period of time (16). Although one unusual African boy has been found with low erythrocyte ADA activity (I2), no immuno- ["C]ADENOSINE INCUBATION logic abnormalities were noted. His lymphocyte enzyme activity was found to be 10% of control (6) and may be high enough to Figure I shows the incubation of control cells with ['"ladeno- sustain grossly normal immunologic function. sine with time. Metabolites of adenosine reach a peak level at 60 The locus for ADA activity is on chromosome 20 (25), and this min, and adenosine levels decrease further at later time points. It gene controls the enzyme activity of all ADA isoenzymes (8). can be seen in Figures 1, 2, and 3 that more [14C]adenosine is lsoenzymes change during immunologic stimulation (9) as enzyme converted to inosine than to adenine nucleotides in control cells at activity increases 2- to 3-fold (7). There are varying amounts of concentrations above 2 x M. Inosine to ATP incorporation residual enzyme activity in enzyme-deficient patients (I). Little is ratios were 2:4 at 2, 3, and 4 x lo-= M adenosine. This ratio known, however, of the role ADA plays in the normal biochemis- increased to 10 at 4 x M adenosine. Figure 2 shows that try of cells. Because decreased ADA activity is almost certainly [14C]adenosine conversion to inosine and hypoxanthine by the causally related to immunologic disease in enzyme-deficient adenosine deaminase reaction is greatly decreased at all levels of patients, we have examined cultured skin fibroblasts in virro for adenosine in ADA-deficient cells. [14C]Adenosine incorporation possible consequences of loss of ADA activity. into GTP as shown in Figure 3 is the same in control and PURINE DYSFUNCTION IN ADENOSINE DEAMINASE DEFICIENCY 64 3 lo-' M adenosine does not inhibit the conversion of I x M ['4C]adenine to ATP. Thus, the decreased [L4C]adenosineconver- sion to ATP found within an experiment at very high adenosine concentrations in ADA-deficient cells may be due to substrate inhibition of adenosine on the enzyme adenosine kinase, and adenosine pool sizes may be larger in these cells. Incorporation of ["Cladenosine into ATP and ADP, as ex- pressed as an ATP:ADP ratio, was also the same in control and ADA-deficient cells. ATP:ADP ratios were 3:l in control and mutant cells at low adenosine concentrations of 2-4 x M. ATP:ADP ratios were higher at higher adenosine concentrations, but were still approximately the same in control and ADA-defi- cient cell strains. .-. CORTROL CELL SlRllRS 21 10 113 0--0 lOIHOSIR1 O~lYlN~SI OIllClERl Clll SIRIIIS 114 111. 110 Fig. 1. Time sequence of adenosine metabolism by control human fibroblasts. Cells were incubated with 4 x M ['%]adenosine. Purine metabolites were separated by thin layer chromatography and high voltage chromatography. 10- 0-0 COHIROL CELL STRllHS 25. 40. 113 B 0--0 IOEROSlRt OflMINISE OlflCIiNT CLLL SIALINS. 154 151 110 0.8 - /* 20 25 . 0-,,, , , 24 810 20 30 I0 I00 lotnoslut COWC~NTRATIONMOLAR . -00 ,--- =- =a4 Fig. 3. Conversion of adenosine to ADP (A) and GTP (B) in control and adenosine deaminase-deficient cell strains. Conditions were the same 151 24 810 20 30 40 100 AOtNOSlNt CONCtNTRlTlON MOLAR . 16- Fig. 2. Conversion of adenosine to inosine (A)and hypoxanthine (B)in 14- ~.COIIIROL CfLL STRllNS. 25. 40. 141 control and mutant cell strains at different concentrations of adenosine. o--OIO~NOSIWIOIIMIN~SI O~~ICILNICILL SIRIIRS. 154 151 110 Incubation was for I hr at 37'. Each point is the average of two determinations. -= 12- -\ Z 2 18- ADA-deficient cells. Adenosine incorporation to ATP and ADP, z as shown in Figure 4, is similar in control and ADA-deficient cell strains except at concentrations of adenosine greater than M. P ['4C]Adenosine conversion to ATP at high adenosine concentra- tions was consistently decreased by about 20% in ADA-deficient 4 6- ---.,- -Cx--Co ,, <----0154 cells at concentrations above 3 x M within any particular experiment, and was statistically significant. However, when data ' 4. -----------o' from five experiments were combined. as shown in Figure 5, there P- ---9:8:-------- was sufficient variation in results so that the differences between -.. mutant and normal cells were not statistically significant. Adeno- /:,---- sine levels required to get decreased levels of ATP formed in !I Ti 8 10 20 ADA-deficient cells, as shown in Figure 5, are 10-fold greater than 30 I0 100 that found toxic to cell growth in Figure 6. It is important to note AO~HOSIH~concrnr~lrlon MOLAR that ATP formation from [14C]adenosine is not increased in Fig. 4. Conversion of adenosine to ATP in control and adenosine ADA-deficient cells. In other experiments not shown, 4,6, and 8 x deaminase-deficient cell strains. Conditions were the same as Figure 2. 644 BENKE AN 5Or a-*COUIROL Ctll SlRIlUS whereas ADA-deficient cells have detached and there are fewer cells than originally plated (Figures 6 and 7). Thus, adenosine is not more toxic to growing control and mutant cells, but does kill nondividing ADA-deficient cells at very high (5 x 10-4-1 x M) adenosine levels. In experiments not shown, the use of fetal calf serum instead of horse serum in toxicity studies greatly decreased the toxicity of adenosine for control and ADA-deficient cell strains. It might be expected that adenosine is more toxic to rapidly proliferating cells, but the rate of cell division is about 24 hr in both lymphocytes and fibroblasts. lshii and Green (I I) have found that cells exposed to toxic amounts of adenosine may be "rescued" with uridine, an end product of pyrimidine metabolism.